Since the late 1950s, the finite element method (FEM) has been an effective tool for solving engineering problems. Initially, calculations were done by hand and based on force rather than displacement. However, with advancements in computer technology, various commercial FEM software programs have been developed, with ANSYS being one of the most popular worldwide, offering a range of products such as Fluent and Workbench.
For beginner Workbench users, selecting the correct meshing method, whether Tetrahedrons, Hexdominant, Sweep, Multizone or Cartesian, can be a challenge in obtaining good mesh quality. But before explaining these methods, it’s important to understand what element types are.
Experienced ANSYS Classic (APDL) users know that selecting the appropriate element type is vital to finite element solutions. In the past, defining the correct element type was often mandatory, typically a 2-dimensional element such as triangular or quadratic. Essentially, selecting the correct element defines the properties of the FE model, such as being a solid block, load-carrying beam, or a flat surface.
Special types of elements are used for different finite element problems to reduce solving time. If you only need to investigate stress on a surface with simple loading, a detailed mesh or element type may not be necessary. It’s better to simplify the model as much as possible. Though we won’t discuss beams, plates, or solid elements in detail, it’s useful to understand what happens in the background of ANSYS Workbench.
In 2D, elements are classified as triangular or quadratic based on their shape and the number of nodes they have, such as TRI3 (3-nodes triangle), TRI6 (6-nodes triangle), QUAD4 (4-nodes quadratic), or QUAD8 (8-nodes quadratic). The higher the number of nodes, the greater the accuracy in strain representation.
2D element shapes
In 3D, elements are 3D versions of 2D elements called tetrahedral and hexahedrons, with varying numbers of nodes such as TET4, TET10, HEX8, HEX20, etc.
3D element shapes
Meshing Methods in Workbench
In Workbench software, you can let the software automatically decide the mesh method based on your boundary conditions or choose from several methods to build your mesh structure. However, it’s important to note that in the user interface (if you’re not using APDL commands), the software doesn’t allow you to choose a specific element such as SOLID186 or SHELL 181. Instead, you control the shape of the elements using meshing methods such as Tetrahedrons, Hex Dominant, Sweep, Multizone, and Cartesian.
In this method, triangular elements are used for creating your mesh, and you may also see prismatic elements in your mesh geometry. Tetrahedrons are preferred for relatively complex geometries such as grooves, channels, and corners with angles. They generate a higher number of cells than quadratic elements with the same mesh size.
Tetrahedral meshes of a ball and a cylinder with a spherical hole
The goal of hex-dominant meshing is to generate meshes dominated by hexahedral elements in both number and volume. However, this method relies on the triangle-merge technique to recombine the initial mesh triangles into quadrilaterals. Therefore, using this method doesn’t mean that all elements will be hexahedrons. You’ll still observe many triangular or prismatic elements in the mesh geometry.
A mixed mesh obtained from an arbitrary tetrahedral mesh. Hexahedra are displayed in purple, prisms in dark blue, pyramids in pink and tetrahedra in light blue. (2) is a cutaway view of (1).
Sweep meshing involves meshing a face of a volume, which is then “swept” through the body to create a volume mesh. The body must have a consistent cross-section to sweep through. For example, a circular face can sweep through a cylinder along its axis.
Multizone is used for meshing complicated single body parts that are too challenging for traditional sweep meshing. This method automatically divides the geometry into mapped (structured/sweepable) regions and free (unstructured) regions. It generates a pure hexahedral mesh where possible, and then fills the more difficult-to-capture regions with unstructured mesh.
The Cartesian cut cell approach generates a mesh without the need for optimization, which removes a significant bottleneck in the simulation pipeline. However, the cut cells created at the boundary can be very small, which imposes a constraint on the explicit stable time step for the simulation. The challenge is to evolve the cut cells using the regular cell time-step without compromising stability and conservation.